课程: Google Cloud Professional Machine Learning Engineer Cert Prep

免费学习该课程！

今天就开通帐号，24,700 门业界名师课程任您挑！

Training models with TensorFlow and a GPU-enabled Docker

Training models with TensorFlow and a GPU-enabled Docker - Google Cloud Platform教程

课程: Google Cloud Professional Machine Learning Engineer Cert Prep

开始一个月试用 购买团队方案

Training models with TensorFlow and a GPU-enabled Docker

“

- [Instructor] Let's take a look at TensorFlow inside of Docker, which is a very interesting way to do machine learning training and also inference. Let's take a look at this website. Docker uses containers to create virtual environments that isolate TensorFlow installation from the rest of the system. So in many scenarios this can work out quite well and in fact you can use the GPU which many people are not aware of. And in fact, if we go through here you can look at some of the instructions here about how to download a TensorFlow Docker image and some of the things that you can do. In fact, there's some great commands here that show you how you can use notebooks or just use the GPU, et cetera. So I've already got this running inside of a GitHub Code Space's environment. In fact, we can take a look at it by just going nvidia-smi l-1 which will show us that it is in fact available, a CUDA version here. We…

内容